PrIINeR: Towards Prior-Informed Implicit Neural Representations for Accelerated MRI
Ziad Al-Haj Hemidi, Eytan Kats, Mattias P. Heinrich · 2025
Read PrIINeR: Towards Prior-Informed Implicit Neural Representations for Accelerated MRI by Ziad Al-Haj Hemidi, Eytan Kats, Mattias P. Heinrich (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-08058.txt.
Description
Ziad Al-Haj Hemidi, Eytan Kats, and Mattias Heinrich speed up MRI scans in 2025 without losing image quality. Shorter scans cut cost and patient strain, yet fast scans blur structure and add ghost patterns. A newer method rebuilds images with a small network fit to each scan, but it fades at high speed-ups for lack of prior knowledge.
Their method, PrIINeR, feeds that per-scan network with knowledge from models pre-trained on many scans. Two checks keep the result true both to the raw scan data and to the prior-informed rebuild. On a public knee and brain data set, it beat its class and several top learning methods, they report. The code is posted in public for others to test.
Catalog record
| CREATOR | Ziad Al-Haj Hemidi, Eytan Kats, Mattias P. Heinrich |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 24 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-08058